Project: Add 8812au rtw88 in-kernel driver. Need testers... #27
Comments
|
Updating to kernel 6.9.3 broke my wifi (I was on aircrack-ng/rtl8812au before), so I tried the repo you linked and it works! System: Pop_OS! Let me know if there is any more info you would like. (I don't know much about drivers, so I don't know what information is pertinent to send). |
|
Adding you to this message to make sure you are on this new issue for the 8812au. Note that this Issue is in the 8812au repo. See first report above.
That is great.
Entering credentials again when you switch ports seems to be normal operations. Now to make sure that the driver is switching to be in the correct mode, can I get you to run the following command while in a usb2 port and then again in a usb3 port: $ lsusb -t Please post both sets of results. On the line for your driver, you should see 5000 if in usb3 mode and 480 if in usb2 mode. We need to make sure that is happening properly.
Well figure out what info we need as we go. You just keep using it and if something does not work like you think it should, let us know what is wrong and we will tell you what info we need. Your help is greatly appreciated. It will be cool when this driver is in mainline. Our 8812au adapters will finally have a proper in-kernel driver and will be plug and play. |
|
Yup, I see 5000 if in usb3 mode and 480 if in usb2 mode. usb3: usb2: |
|
The rtw8812au driver works flawlessly with my (previously useless) ALFA AWUS036ACH! Monitor mode works! (see screenshot) The driver also works with USB 3 and USB 2. USB 2 Port: |
|
Finally we get the kernel driver! Thank you! Model: ALFA AWUS036ACH
|
|
@kikitte Someone sent a fix for those warnings: https://lore.kernel.org/linux-wireless/[email protected]/ @Jake-Grafton What is the application in your screenshot? |
|
Currently working with this driver. Installed correctly, but can't get this to go into monitor mode for whatever reason. Card being used is an Alfa AWUS036ACH, realtek 8812au driver. sudo ip link set wlan1 down
sudo iw wlan1 set monitor none
command failed: Operation not supported (-95)This will connect to Wi-Fi and hold a connection fine, but monitor mode will error out. UPDATE: tried a clean install of the experimental driver and nothing. Going to keep trying and update as able. |
UPDATE: Monitor mode works with aircrack-ng but not working with hcxdumptool. I am using the wrong command to test hcxdumptool. |
hcxdumptool, as it is vastly superior to the aircrack-ng suite. |
|
The author of hcxdumptool is @ZerBea . He was helping with the testing of 8821/11au. He can likely answer questions about hcxdumptool if you have any. We may need to see if anyone has an extra 8812au based adapter that we can get into his hands. Unfortunately I only have one 8812au based adapter or I would get one in the mail to him today. |
|
@morrownr I'll order this adapter: According to Delock, it should be a RTL8812AU based one: There is no need to run hcxdumptool on high TX power devices. |
|
@morrownr The |
|
How about this? |
|
I've noticed there's ongoing development of a mac80211 driver for the RTL8812AU chip. As someone who's been following this topic casually, I'm hoping for some insights:
I've attempted to research these questions independently but haven't found comprehensive answers. As a novice in this field, I'd greatly appreciate any insights from those more knowledgeable about the situation. Thank you for any information or clarification you can provide! |
|
Hi @sisi399
You are at the right place. Read message one in this thread for instructions on how to test.
Larry Finger passed away last Friday. RIP Larry. He will be greatly missed but we are trying to carry on without him. We are still using his rtw88 repo to do the work. @dubhater and I have the appropriate rights to use the site. I am really busy right now so maybe we can get to more of your questions later. Right now, we need testers so please read the first message in thread and consider helping us by testing. Regards, |
|
AP mode should work now. I also enabled power saving. @castillofrancodamian You will be disappointed: As far as I can tell, it's because of the way rtw88 decides when to enter power saving and for how long. The official driver makes those decisions differently. Something to ask Ping-Ke about... |
|
@dubhater |
|
Is that the same as running |
Maybe your rtl8812au driver was installed via dkms, you can check it with this command. |
|
@dubhater I have a question: I checked uninstall step in Makefile, won't the system crash |
|
@dubhater @morrownr : I tried 4a1ee64475672a9186e10d87ac5e251448f86914 (beacon fix) with a 8822bu device. In my case, unfortunatly i still can't see my hotspot on 5GHz band using an intel wifi card (windows 10) and also my lenovo duet device (chromeos). Using realtek driver, i have no issue. My phone (realtek) and my android box (mediatek) can both see it (for these two devices hotspot seems to work). Not sure your fix fix beacon issue for 8822bu device unfortunatly :'( |
@morrownr I thought one of the pull requests I merged recently broke it, but no, that part of the Makefile hasn't changed. It actually looks like a bug in modprobe. From its manual:
@a5a5aa555oo I'm unloading rtw_8822ce all the time and nothing bad happens. @castillofrancodamian You don't need to uninstall rtw88, or to install 8812au. It's enough to manually unload rtw_8812au (the others don't matter for this case), then you can load 8812au. Works until you reboot. rmmod rtw_8812au
insmod /path/to/8812au.ko rtw_switch_usb_mode=1 |
|
Or maybe one of those changes did break it. The SDIO drivers should be uninstalled only if they were compiled. |
I made a patch to fix it, I'll open a pull request later. diff --git a/Makefile b/Makefile
index c97133b..bbec8ec 100644
--- a/Makefile
+++ b/Makefile
@@ -184,35 +184,12 @@ endif
@echo "The driver rtw88 and its firmware installed successfully!"
uninstall:
- modprobe -r rtw_8723cs
-ifeq ($(CONFIG_PCI), y)
- modprobe -r rtw_8723de
-endif
- modprobe -r rtw_8723ds
- modprobe -r rtw_8723du
- modprobe -r rtw_8812au
- modprobe -r rtw_8821au
-ifeq ($(CONFIG_PCI), y)
- modprobe -r rtw_8821ce
-endif
- modprobe -r rtw_8821cs
- modprobe -r rtw_8821cu
-ifeq ($(CONFIG_PCI), y)
- modprobe -r rtw_8822be
-endif
- modprobe -r rtw_8822bs
- modprobe -r rtw_8822bu
-ifeq ($(CONFIG_PCI), y)
- modprobe -r rtw_8822ce
-endif
- modprobe -r rtw_8822cs
- modprobe -r rtw_8822cu
-
- rm -f $(MODDESTDIR)/rtw_*.ko*
-
+ @-modprobe -raq rtw_8723{cs,de,ds,du}
+ @-modprobe -raq rtw_8822{be,bs,bu,ce,cs,cu}
+ @-modprobe -raq rtw_8821{au,ce,cs,cu} rtw_8812au
+ @rm -vf $(MODDESTDIR)/rtw_*.ko*
depmod $(DEPMOD_ARGS)
-
- @echo "Uninstall rtw88 SUCCESS"
+ @echo "The driver rtw88 was uninstalled successfully." |
That's how it is. When I tested the rtl8812au driver I did it using dkms.
I get: |
|
If you don't want rtl8812au driver be built/installed during a kernel update, you can remove it. |
The rest of the driver remains unremoved. After reboot. I think I'm going to have to use the old reliable: reinstall the operating system. |
|
Before reinstalling your OS, try just removing /var/lib/dkms/rtl8812au-5.13.6
|
It appears to have been removed. |
|
Another way to correct that problem: sudo apt purge dkms That is for Debian but the idea is to uninstall it. You can always reinstall it if you actually need it. |
I reinstalled rtl8812au (after removing rtw_8812au) using dkms as asked by dubhater in this comment, so I don't want to remove dkms at the moment. |
I understand but the 8812au driver in this repo can be installed without dkms: $ make Just trying to give you an option to clean up that specific dkms issue. |
|
@morrownr I already installed rtl8812au manually and I see the same symptoms as with Installation
|
|
@castillofrancodamian That's a relief. :) Are you using tlp ? |
What's that? A tool? |
|
Something that causes problems: https://wiki.archlinux.org/title/TLP |
|
I've been sick for 3 days but will test AP mode on rtw_8812au as a priority as able. |
|
Hi @dubhater , i could test again rtw88 driver on : lwfinger/rtw88@4a1ee64 |
|
On some devices (ex : my phone) connexion continue to work even when beacons stop. Please note that beacon and drv rsvd page failure do not always happen (whatever if the beacons are working or not) |
|
Using airdump, we can clearly see no beacons are emitted on channel 36 currently, eventhough hotspot is working on some devices 30:93:BC:9F:C0:D5 -77 10 4 0 100 1733 WPA2 CCMP PSK Livebox-C0D0 BSSID STATION PWR Rate Lost Frames Notes Probes 30:93:BC:9F:C0:D5 20:1F:3B:83:81:53 -72 0 - 6 0 1 |
My PC is a desktop. Therefore it is not an energy problem. |
This should be fixed now. |
|
@dubhater i confirm that lwfinger/rtw88@610e04f fixed definitively beacon issues on 8822bu. Well done, thank you for all your team work, this is really impressive ! [ ID] Interval Transfer Bitrate Retr iperf Done. In normal times, i have around 200mb bandwidth, so something is still wrong. But what a progress ! |
|
@morrownr i wouls say rtl8822bu, as i'm loading the driver using : Here is are some logs, using the realteck driver. sudo lsusb -t I only have usb 2.0, but even for 2.0, bandwith is adnormally low compared to realtek driver |
|
"failed to get urb length:32768" and "Rate marked as a VHT rate but data is invalid" errors are fixed now. |
Greetings to anyone that reads this message.
This Issue is where we coordinate and take bug reports for the new 8812au in-kernel driver.
An effort is underway to add support for the rtl8812au chipset to the rtw88 in-kernel driver series. The driver is available for testing at the following repo:
https://github.com/lwfinger/rtw88
Remember to first remove the out-of-kernel driver in this repo or whatever repo you may have installed. You can run the following to remove it if using this repo:
$ sudo sh remove-driver.sh
It is important to follow the instructions in the README at the repo with the new test driver. You may not be familiar with rtw88 in the kernel but even if you are, there are some necessary mods that you need to know about. The rtw88 that has this new driver is more advanced than the rtw88 in stable kernels as it follows wireless-next and is used to work on and develop new drivers.
We welcome you to test and report on this new driver. Your testing will help us get this new driver in the Linux kernel sooner and in better shape. If you are aware of anyone who is familiar with mac80211 drivers, please invite them as more eyes on the code is a good thing. Your ideas are most welcome. We can do this.
@morrownr
Status report:
Reports go in this issue.
The text was updated successfully, but these errors were encountered: